JP7394023B2 - Welding work evaluation device, welding work evaluation method and program - Google Patents

Description

The present invention relates to a welding work evaluation device, a welding work evaluation method, and a program.

Until now, experienced welders have spent time educating beginners in advanced skills such as welding through on-the-job training and other means. However, with the rapid decline in the number of skilled welding workers due to the declining birthrate and aging population in recent years, beginners may not necessarily be able to receive training from experts, and skilled skills are rapidly disappearing from manufacturing sites. Under these circumstances, efforts are being made to automate welding work using robots, which have achieved results in reducing man-hours and stabilizing quality. On the other hand, there is still a strong need for manual welding in narrow spaces that cannot be accessed by robots or in welding large structures with large dimensional variations, and there is a need for a system that can efficiently train the next generation of workers.

In view of this situation in recent years, several support systems and methods for welding work education have been proposed. For example, the summary of Patent Document 1 below states, ``A welding skill education support system that can easily simulate the welding environment and can educate, correct, or instruct welding work even if the student is self-studying. provided.”
In addition, the summary of Patent Document 2 below states, "We provide a quality experience training device and a quality experience training method that allow trainees to easily simulate nonconformity cases through a model that reproduces nonconformity cases." ing.

JP2013-156428A Japanese Patent Application Publication No. 2017-111387

By the way, in the above-mentioned technology, there is a desire to realize more appropriate skill evaluation of welding work.
The present invention has been made in view of the above-mentioned circumstances, and it is an object of the present invention to provide a welding work evaluation device, a welding work evaluation method, and a program that realize more appropriate skill evaluation of welding work.

In order to solve the above problems, the welding work evaluation device of the present invention includes a coordinate measuring unit that acquires a torch position, which is the position of a welding torch during welding work, and a coordinate measurement unit that measures the movement of the torch position from among a plurality of work evaluation algorithms. an algorithm selection section that selects one of the work evaluation algorithms to perform the welding work; and a welding work evaluation section that evaluates the welding work based on the selected work evaluation algorithm; The selected work evaluation algorithm is determined by selecting at least two of the reference data from a plurality of reference data each corresponding to a part of the work evaluation algorithm, and interpolating the selected at least two reference data. The present invention is characterized in that it includes a complementing section that generates new reference data corresponding to .

According to the present invention, more appropriate skill evaluation of welding work can be realized.

FIG. 1 is a schematic diagram of a welding work evaluation system according to a first preferred embodiment. FIG. 2 is a schematic diagram of a welding work evaluation system according to a second preferred embodiment. It is a flowchart of welding work evaluation processing performed in a control device. FIG. 3 is a diagram showing the principle of estimating the welding posture of a welding worker. FIG. 3 is a diagram showing an example of a welding work evaluation section. It is a flowchart of a scoring process subroutine. FIG. 3 is a schematic diagram of a welding work evaluation system according to a third preferred embodiment.

[Premise of embodiment]
Welding work is a complex work that involves a combination of many physical phenomena, and skilled welders fine-tune welding conditions based on their extensive experience. For example, even if the welding method is the same, the way gravity acts on the molten metal is different when welding is performed in an upward position while looking up at the joint and when welding is performed in a downward position while looking down at the joint. Generally speaking, when welding in an upward position, the weld metal tends to drip during welding, so it is necessary to increase the welding speed and solidify it quickly.

In the technique to which Patent Document 1 described above is applied, welding postures are limited, and skilled skills cannot be sufficiently trained. Another problem is that it is difficult to evaluate welding work in which the welding position gradually changes, such as when welding a cylinder. In addition, the technology applying Patent Documents 1 and 2 does not assume cases where there are large variations in dimensions, and can only teach welding patterns that achieve ideal assembly. It cannot be said that this method can train welding workers who are required to respond flexibly.

Therefore, in the preferred embodiment described below, a system is provided that can evaluate welding work by applying appropriate reference data and work evaluation algorithms to each welding position, even when the welding positions are different. do.

[First embodiment]
FIG. 1 is a schematic diagram of a welding work evaluation system 1 according to a first preferred embodiment.
In FIG. 1, a welding work evaluation system 1 includes a signal transmitter 11, a signal receiver 12, a digital welder 60, a welding torch 62, a control device 100 (welding work evaluation device), a display 140, and an input device. 142. The input device 142 includes a mouse, a keyboard, etc. (not shown), and the operating states thereof are supplied to the control device 100. The display 140 displays various information under the control of the control device 100. Furthermore, devices such as the digital welding machine 60, the signal receiving section 12, and the control device 100 are connected to each other via a communication network such as a wireless LAN or a wired LAN.

The digital welding machine 60 has a function of managing the current and voltage supplied to the welding torch 62, the amount of welding material supplied, and the like. Further, the current, voltage, supply amount of welding material, etc. are supplied to the control device 100. Thereby, the control device 100 can display the current, voltage, supply amount of welding material, etc. on the display 140 at any time. Welding torch 62 is held by welding operator 50 . When power is supplied from the digital welding machine 60 to the welding torch 62, the material to be welded 70 is thereby welded, and a weld bead 72 is formed on the material to be welded 70.

The signal transmitting unit 11 is attached to the welding torch 62 and calculates the coordinates of the tip position of the welding torch 62 (hereinafter referred to as the torch position) and the inclination angle of the central axis of the welding torch 62 with respect to the horizontal plane (hereinafter referred to as the torch inclination angle). , transmits coordinate information that changes according to. For example, a plurality of reflective material coated markers, a GPS sensor, a horizontal sensor, an inertial sensor, etc. that specify position coordinates by reflecting a specific wavelength can be applied to the signal transmitter 11. The signal receiving section 12 receives the coordinate information transmitted from the signal transmitting section 11 and supplies this coordinate information to the control device 100.

For example, a motion capture device or a receiver capable of receiving signals from the various sensors described above can be applied to the signal receiving unit 12. In actual welding, strong arc light is generated, so it is desirable to use a transmitter and a receiver that are less susceptible to this influence for the signal transmitting section 11 and the signal receiving section 12. In the illustrated example, the signal transmitting section 11 is attached to the welding torch 62, but the signal transmitting section 11 may be attached to, for example, gloves worn by the welding operator 50.

The control device 100 is equipped with general computer hardware such as a CPU (Central Processing Unit), a RAM (Random Access Memory), a ROM (Read Only Memory), and an SSD (Solid State Drive). , an OS (Operating System), application programs, various data, etc. are stored therein. The OS and application programs are loaded into RAM and executed by the CPU. In FIG. 1, inside the control device 100, functions realized by application programs and the like are shown as blocks.

That is, the control device 100 includes a coordinate measurement section 110, an algorithm selection section 120, and a welding work evaluation section 130. The coordinate measurement unit 110 (coordinate measurement process, coordinate measurement means) includes an acquisition unit 113 and a calculation unit 114. Further, the algorithm selection section 120 (algorithm selection process, algorithm selection means) includes a work evaluation database 121, a comparison selection section 122, and a posture estimation section 123.

Further, the welding work evaluation unit 130 (welding work evaluation means, welding work evaluation process) includes a reference database 131, a comparison unit 132, an evaluation unit 133, a database complementation unit 134 (complementation unit), and a display control unit 135. , is equipped with.

The coordinate measurement unit 110 calculates the torch position and torch inclination angle. The algorithm selection unit 120 selects one of the plurality of work evaluation algorithms based on the torch position and the torch inclination angle. The welding work evaluation unit 130 evaluates the welding work performed by the welding operator 50 based on the selected work evaluation algorithm.

In the coordinate measurement section 110, an acquisition section 113 receives the coordinate information of the welding torch 62 from the signal reception section 12 and temporarily stores it. The calculation unit 114 calculates the torch position and torch inclination angle based on the coordinate information, and records the results as "welding state data" in association with time information. The calculated welding state data is supplied to the algorithm selection section 120 and the welding work evaluation section 130.

The work evaluation database 121 stores in advance a plurality of work evaluation algorithms corresponding to ranges of torch positions and torch inclination angles. Posture estimating section 123 estimates the welding posture of welding operator 50 based on the torch position and the range of the torch inclination angle. Then, the comparison selection unit 122 selects one of the work evaluation algorithms based on the estimated welding posture (estimated welding posture). However, the comparison selection unit 122 may select any work evaluation algorithm based on the torch position and torch inclination angle included in the welding state data, without using the estimated welding posture.

In the welding work evaluation unit 130, a reference database 131 stores reference data corresponding to the range of the torch position and the torch inclination angle. Here, the reference data is data indicating a reference welding work state, for example, an ideal welding work state. The comparison unit 132 compares the actual welding operation with reference data and calculates the difference between the two. The evaluation unit 133 evaluates the welding work of the welding operator 50 based on the difference output by the comparison unit 132.

In addition, the database complementing unit 134 performs a process of determining other torch positions and torch inclination angle ranges for which reference data is not recorded in the reference database 131 (in other words, for welding postures that are not recorded in the database). Compute the reference data by interpolating the reference data. In other words, the database complementation unit 134 selects at least two pieces of reference data from among a plurality of pieces of reference data each corresponding to a part of a plurality of work evaluation algorithms, and interpolates the selected at least two pieces of reference data. By doing so, new reference data corresponding to the selected work evaluation algorithm is generated.

However, if the reference database 131 is complete, the database complementing section 134 can be omitted. The display control unit 135 causes the evaluation result by the evaluation unit 133 to be displayed on an external display. Further, the display control unit 135 may display the current value and voltage value that the digital welding machine 60 is supplying to the welding torch 62 on the display.

Further, instead of the reference database 131 and the comparison unit 132, a learning model that has learned the movements of a plurality of skilled welding operators through machine learning may be applied. In this case, the learning model outputs, for example, the degree of deviation from ideal operation when welding state data during actual welding is input. Then, the evaluation unit 133 may evaluate the welding work of the welding operator 50 according to this degree of deviation.

Further, the display connected to the display control unit 135 may be one that displays information to a person outside the welding booth, for example. Alternatively, it may be a small display mounted inside the welding surface 52 worn by the welding operator 50. When the latter display is applied, the evaluation results by the evaluation unit 133 can be fed back to the welding operator 50 one by one.

[Second embodiment]
FIG. 2 is a schematic diagram of a welding work evaluation system 2 according to a second preferred embodiment. In the following description, parts corresponding to those in the first embodiment described above may be denoted by the same reference numerals, and the description thereof may be omitted.
In FIG. 2, the material to be welded 70 is a cylindrical test piece 170 formed in a cylindrical shape, and is arranged horizontally. A plurality of shielding plates 160 each having a substantially rectangular plate shape are arranged around the cylindrical test piece 170, and a welding booth 180 is formed by these shielding plates 160.

Therefore, the control device 100 and the display 140 connected thereto are installed outside the welding booth 180. This prevents arc light and spatter from reaching the control device 100 and the display 140. A motion capture device 152 is placed on the side of the welding worker 50. A plurality of reflective material coating markers 64 (markers) are attached near the tip of the welding torch 62 as the signal transmitter 11 . By attaching a plurality of reflective material coating markers 64 in this way, the movement of the tip of the welding torch 62 can be tracked in detail.

Motion capture device 152 detects the postures and movements of welding operator 50 and welding torch 62 based on the reflected light from reflective material application marker 64, and thereby specifies the position and inclination of welding torch 62. In addition, approximately rectangular through holes 162 and 164 are formed at various locations in the shielding plate 160, and motion capture devices 150 and 154 are arranged outside of these through holes 162 and 164. These motion capture devices 150, 152, and 154 function as the signal receiving section 12. In this way, by applying a plurality of motion capture devices 150, 152, 154, the postures and movements of welding worker 50 and welding torch 62 can be detected with high precision.

In FIG. 2, the axial direction of the cylindrical test piece 170 is defined as the z-axis. Further, the horizontal axis perpendicular to the z-axis is the x-axis, and the vertical axis perpendicular to the x-axis and the z-axis is the y-axis. A jig (not shown) for fixing the cylindrical test piece 170 is arranged inside the welding booth 180. Thereby, the position and orientation of the cylindrical test piece 170 are kept constant, and even if the cylindrical test piece 170 is replaced, there is no need to change the coordinate system of the x-axis, y-axis, and z-axis.

The configuration of the control device 100 is similar to that of the first embodiment (see FIG. 1). That is, the coordinate measuring section 110 receives the coordinate information of the welding torch 62 from the motion capture devices 150, 152, and 154, which are the signal receiving section 12, and calculates the torch position and torch inclination angle. Then, the coordinate measurement section 110 supplies the calculated torch position and torch inclination angle to the algorithm selection section 120 and the welding work evaluation section 130 at any time.

FIG. 3 is a flowchart of welding work evaluation processing executed by control device 100.
When the process proceeds to step S1 in FIG. 3, the control device 100 starts welding work evaluation when the welding operator 50 starts the work. However, step S1 may be started slightly before the welding operation. Alternatively, step S1 may be automatically executed in response to an arc ignition start signal supplied from the digital welding machine 60.

Next, when the process advances to step S2, a "welding work evaluation section" is input. That is, when the user operates input device 142 to input a welding work evaluation section, control device 100 stores the contents. Here, the welding work evaluation section refers to a section in which the control device 100 evaluates the welding work, and if it is desired to evaluate the entire welding work, it is sufficient to evaluate the period from arc ignition to the time of extinction. Furthermore, if it is desired to evaluate the welding work in a part of the section in detail, it is preferable to input the welding work evaluation section for that range. Further, the "welding work evaluation section" may be specified by a "time range", or an evaluation section may be provided at fixed time intervals. For example, in the cylindrical coordinate system, the range from 0° to 30° can be set as the welding work evaluation section.

Next, in step S3, welding work is started. In the welding operation, the welding material is melted by arc heat, and a bead 172 is formed on the cylindrical test piece 170, as shown in FIG. At this time, the current, voltage, supply amount of welding material, etc. supplied to the welding torch 62 are supplied from the digital welding machine 60 to the control device 100 at any time.

Next, when the process proceeds to step S4, the control device 100 detects the torch position and torch inclination angle. That is, motion capture devices 150 to 154 detect the positions of a plurality of reflective material application markers 64 attached to welding torch 62, and based on the results, coordinate measurement unit 110 calculates the torch position and torch inclination angle. .

More specifically, first, the coordinate measurement unit 110 (see FIG. 1) associates coordinate information with time information and records the result as primary data. The calculation unit 114 (see FIG. 1) calculates the torch position and torch inclination angle at any time based on this primary data, associates the calculated torch position and torch inclination angle with time information, and calculates the torch position and torch inclination angle with the secondary data. It is recorded as a certain "welding state data". Further, the torch position and torch inclination angle are displayed on the display 140 at any time, and people other than the welding operator 50 can also observe the welding state data during the welding operation.

Next, when the process proceeds to step S5, the coordinate measuring unit 110 determines whether the torch position has reached the end position of the welding work evaluation section. Note that when the welding work evaluation section is specified by a "time range", the coordinate measurement unit 110 determines whether the current time has reached the end time. If the determination in step S5 is "no", the process in step S4 is repeated. On the other hand, if the determination in step S5 is "yes", the process proceeds to step S6. In step S6, the coordinate measurement unit 110 estimates the welding posture of the welding worker 50 based on the torch position and torch inclination angle at the start position and end position of the welding work evaluation section. The details will be explained with reference to FIGS. 4 and 5.

FIG. 4 is a diagram showing the principle of estimating the welding posture of the welding operator 50.
In FIG. 4, the upward angle along the y-axis is 0°. Here, the torch position, that is, the tip position of the welding torch 62 is located along the outer periphery of the cylindrical test piece 170, so the torch position can be expressed by the angle with respect to the y-axis. This angle is called the torch polar coordinate angle θ. If the torch polar coordinate angle θ is in the range of -θ1 to θ1, it is considered that the welding operator 50 performs the welding operation in a posture that looks down on the cylindrical test piece 170. Therefore, this posture is called the "downward posture."

Further, if the torch polar coordinate angle θ is in the range of θ2 to -θ2, it is considered that the welding operator 50 performs the welding operation in a posture that looks up at the cylindrical test piece 170 from below. Therefore, this posture is called the "upward posture." Further, if the torch polar coordinate angle θ is in the range of θ1 to θ2 or −θ1 to −θ2, it is considered that the welding operator 50 performs the welding operation in a posture standing against the cylindrical test piece 170. Therefore, this posture is called the "standing posture."

FIG. 5 is a diagram showing an example of welding work evaluation sections.
For example, if the welding work evaluation section is in the range of "torch polar coordinate angle 0° to 30°", position P1 of the welding torch 62 becomes the starting position of the welding work evaluation section, and position P2 becomes the ending position.

The comparison selection unit 122 (see FIG. 1) of the algorithm selection unit 120 described above calculates the torch polar coordinate angle θ and torch inclination angle at the start position of the welding work evaluation section, and the torch polar coordinate angle θ and torch inclination angle at the end position. compare. For example, as shown in FIG. 5, if the moving distance of the tip position of the welding torch 62 in the x-axis direction (horizontal direction) is larger than the moving distance in the y-axis direction (vertical direction), the welding torch 62 mainly moves in the horizontal direction. It can be concluded that it did.

Further, as shown in FIG. 5, if the welding torch 62 is inclined at a positive angle of about 45° to 75° with respect to the x-axis direction, the welding operator 50 can It is presumed that the welding is done facing downwards. Therefore, in the above example, the comparison selection unit 122 determines that the estimated welding posture of the welding worker 50 in the welding work evaluation section is a downward posture (see FIG. 4) based on both the torch position and the torch inclination angle. judge.

Further, if the moving distance of the tip position of the welding torch 62 in the x-axis direction is larger than the moving distance in the y-axis direction, and the welding torch 62 is inclined by a negative angle with respect to the x-axis direction, welding It is presumed that the worker 50 is welding the cylindrical test piece 170 facing upward at the welding location. Therefore, the comparison selection unit 122 determines that the estimated welding posture of the welding worker 50 in the welding work evaluation section is an upward posture (see FIG. 4).

Further, if the moving distance of the tip position of the welding torch 62 in the y-axis direction is larger than the moving distance in the x-axis direction, the welding operator 50 may be welding while standing against the welding location of the cylindrical test piece 170. Presumed. Therefore, the comparison selection unit 122 determines that the estimated welding posture of the welding worker 50 in the welding work evaluation section is the vertical posture. Furthermore, the comparison selection unit 122 determines that if the y-coordinate value of the start position is larger than the y-coordinate value of the end position, it is the vertical downward movement posture, and if it is smaller, it is the vertical movement posture. The comparison and selection section 122 transmits the determined estimated welding posture and the work evaluation algorithm corresponding to the estimated welding posture to the welding work evaluation section 130.

Returning to FIG. 3, when the process proceeds to step S7, the welding work evaluation unit 130 checks the reference database 131 and scores the welding state data in accordance with the determined estimated welding posture. Specifically, in step S7, the scoring subroutine shown in FIG. 6 is called. Note that the "lateral posture" shown in FIG. 3 will be described in detail in the third embodiment described later.

FIG. 6 is a flowchart of the scoring subroutine (FIG. 6) called in step S7.
When the process proceeds to step S71 in FIG. 6, the welding work evaluation unit 130 starts scoring processing. Next, when the process advances to step S72, the comparison unit 132 selects an evaluation parameter. Here, the evaluation parameter is a parameter that affects welding quality. Parameters that are candidates for evaluation parameters include, for example, welding speed, torch inclination angle, weaving speed, toe stop time, welding current, welding voltage, wire supply amount, and the like. Here, as an example, a case will be described in which welding speed and torch inclination angle, which are considered to have a strong influence on welding quality, are selected as evaluation parameters.

Next, when the process proceeds to step S73, the comparison unit 132 reads reference data corresponding to the previously identified estimated welding posture from the reference database 131. For example, assume that the specified estimated welding posture is a "downward posture," the welding speed in the reference data is 200 mm/min, and the torch inclination angle in the reference data is 20 degrees.

Next, when the process proceeds to step S74, the comparison unit 132 calculates the difference between the ideal motion in the reference data and the actually acquired welding state data. For example, if the average welding speed in the welding state data is 180 mm/min, the difference from the reference data (200 mm/min) is "±10%". Further, if the average torch inclination angle in the welding state data is 15°, the difference from the reference data (20°) is “±25%”.

Next, when the process proceeds to step S75, the comparison unit 132 determines whether or not the sum of these difference values is to be used as the evaluation result. If it is determined "Yes" here, the process proceeds to step S76, and the evaluation unit 133 evaluates each parameter (welding speed, torch inclination angle, weaving speed, toe stop time, welding current, welding voltage, wire supply amount etc.), weighting processing is performed on the difference values, the results are added up, and this is used as the scoring result. Note that the designation of whether or not to add up the difference values and the weighting of each parameter in the case of adding up are included in the work evaluation algorithm included in the work evaluation database 121 described above.

For example, if the work evaluation algorithm specifies that "welding speed and torch inclination angle are weighted together at 4:1," then "the difference is ±13% with respect to the ideal welding motion." You can get the following scoring result. Note that if the determination in step S75 is "no", the difference value for each parameter becomes the scoring result as it is. Next, when the process proceeds to step S77, the scoring ends and the process returns to the welding work evaluation process (FIG. 3).

In FIG. 3, when the process next advances to step S8, the display control unit 135 (see FIG. 1) causes the display 140 to display the above-mentioned scoring results and the like. In addition to the scoring results, the display 140 may display primary data such as torch trajectory, secondary data such as welding speed and torch angle, and current and voltage values obtained from digital welding machine 60. Next, when the process proceeds to step S9, the control device 100 determines whether or not the welding work has been completed.

Here, if the determination is "no", the process returns to step S4, and the processes of steps S4 to S8 described above are repeated for the next welding work evaluation section. If the determination in step S9 is "yes", the process proceeds to step S10, and the control device 100 ends the welding work evaluation process.

[Third embodiment]
FIG. 7 is a schematic diagram of a welding work evaluation system 3 according to a third preferred embodiment. In the following description, parts corresponding to those in the first embodiment described above may be denoted by the same reference numerals, and the description thereof may be omitted.
In FIG. 7, cylindrical test piece 170 is placed on the floor of welding booth 180 so that its axial direction is the y-axis direction.

Both the x-coordinate value and the z-coordinate value of the center of the circle of the cylindrical test piece 170 are set to "0", and the y-coordinate value of the upper end surface of the cylindrical test piece 170 is set to "0". Also in this embodiment, a jig (not shown) for fixing the cylindrical test piece 170 is arranged inside the welding booth 180. Thereby, the position and orientation of the cylindrical test piece 170 are kept constant, and even if the cylindrical test piece 170 is replaced, there is no need to change the coordinate system of the x-axis, y-axis, and z-axis. In this embodiment, welding work is performed along the circumferential direction of the cylindrical test piece 170.

Further, in the shielding plate 160 of this embodiment, a through hole 166 is formed at a position horizontally opposite to the cylindrical test piece 170, and a motion capture device 156 is installed outside the through hole 166. Thereby, the motion capture device 156 can reliably capture the position of the reflective material application marker 64. The configuration of the welding work evaluation system 3 other than that described above is the same as that of the second embodiment (see FIG. 2).

Next, the operation of this embodiment will be explained. The welding work evaluation process (FIG. 3) is also executed in the control device 100 of this embodiment.
In FIG. 3, the processing of steps S1 to S5 is similar to that of the second embodiment. That is, the control device 100 starts welding work evaluation in step S1, a welding work evaluation section is input in step S2, and welding work is started in step S3. Next, in step S4, control device 100 detects the torch position and torch inclination angle, and in step S5, coordinate measuring section 110 determines whether the torch position has reached the end position of the welding work evaluation section.

Next, in step S6, the coordinate measuring unit 110 specifies the estimated welding posture of the welding worker 50 based on the torch position and torch inclination angle at the start position and end position of the welding work evaluation section. As shown in FIG. 7, in this embodiment, the welding operator 50 forms a weld bead 172 along the circumferential direction of the cylindrical test piece 170.

Here, if the welding work evaluation section is, for example, a range of 90° to 120° of the torch polar coordinate angle θ, the x and z coordinate values of the torch position change greatly in the welding work evaluation section, and the y coordinate value remains almost unchanged. In such a case, the welding operator 50 is considered to be welding the cylindrical test piece 170 while facing sideways to the welding location. Therefore, in the above example, the comparison selection unit 122 determines that the estimated welding posture of the welding worker 50 in the welding work evaluation section is a horizontal posture based on both the torch position and the torch inclination angle.

In FIG. 3, when the process next advances to step S7, the scoring process subroutine (FIG. 6) is called, as in the second embodiment. In this routine, the processing in steps S71 to S73 is the same as that in the second embodiment described above. That is, in step S71, welding work evaluation section 130 starts scoring processing, and in step S72, comparison section 132 selects evaluation parameters.

Next, when the process proceeds to step S73, the comparison unit 132 reads from the reference database 131 reference data corresponding to the previously identified estimated welding orientation, that is, the horizontal orientation. In this horizontal position, it is assumed that the welding speed in the reference data is 150 mm/min and the torch inclination angle in the reference data is 30°.

Next, when the process proceeds to step S74, the comparison unit 132 calculates the difference between the ideal motion in the reference data and the actually acquired welding state data. Here, it is assumed that the average welding speed was 180 mm/min and the average torch inclination angle was 30° in the actually acquired welding state data. Then, the difference in average welding speed becomes "±20%" and the difference in torch inclination angle becomes "0".

Next, the processing in steps S75 to S77 is also similar to that in the second embodiment. That is, in steps S75 to S77, the difference values of individual parameters or the results of weighting these difference values and adding them up are used as the scoring results based on the horizontal posture work evaluation algorithm. For example, if the work evaluation algorithm for horizontal position stipulates that the welding speed and torch inclination angle are weighted 4:1, then the difference is ±16% with respect to the ideal welding motion. ” is obtained. Next, the process returns to the welding work evaluation process (FIG. 3), and the processes in steps S8 to S10 here are also similar to those in the second embodiment.

As described above, according to the second and third embodiments, even if the estimated welding postures of the welding operator 50 are different, the welding can be performed by referring to appropriate standard data for each estimated welding posture. It is possible to evaluate work and efficiently train the next generation of welding workers.

[Effects of embodiment]
According to the preferred embodiment as described above, the welding work evaluation device (100) includes a coordinate measuring section (110) that acquires the torch position that is the position of the welding torch (62) in the welding work, and a plurality of work evaluations. An algorithm selection unit (120) that selects one work evaluation algorithm corresponding to the movement of the torch position from among the algorithms, and a welding work evaluation unit (130) that evaluates the welding work based on the selected work evaluation algorithm. and. Thereby, a work evaluation algorithm corresponding to the movement of the torch position can be adopted from among a plurality of work evaluation algorithms, and a more appropriate skill evaluation of welding work can be realized.

Here, the welding torch (62) or the welding worker (50) operating the welding torch (62) is equipped with a marker (64), and the coordinate measurement unit (110) measures the position of the marker (64). It is even more preferable to have the function of calculating the torch position and the torch inclination angle, which is the inclination of the welding torch (62), based on information from the motion capture device (150-156). This allows the torch inclination angle to be measured, making it possible to achieve more appropriate skill evaluation.

Furthermore, the algorithm selection unit (120) estimates the posture of the welding worker (50) operating the welding torch (62) based on the torch position and the torch inclination angle, which is the inclination of the welding torch (62). It is more preferable to include a posture estimation section (123) and a comparison selection section (122) that selects a work evaluation algorithm corresponding to the estimated welding posture that is the estimation result of the posture estimation section (123). Thereby, a work evaluation algorithm corresponding to the estimated welding posture can be selected, and more appropriate skill evaluation can be realized.

Further, the welding work evaluation unit (130) selects one reference data corresponding to the selected work evaluation algorithm from among the plurality of reference data corresponding to the plurality of work evaluation algorithms, and combines the selected reference data with the selected reference data. It is more preferable to include a comparison unit (132) that compares the movement of the torch position with the movement of the torch position. This makes it possible to select appropriate reference data, thereby realizing more appropriate skill evaluation.

Further, the welding work evaluation unit (130) selects at least two reference data from among the plurality of reference data each corresponding to a part of the plurality of work evaluation algorithms, and interpolates the selected at least two reference data. The present invention includes a complementing unit (134) that generates new reference data corresponding to the selected work evaluation algorithm, and a comparison unit (132) that compares the new reference data with the movement of the torch position. More preferred. As a result, even when there are only a few types of reference data prepared in advance, appropriate skill evaluation can be achieved.

[Modified example]
The present invention is not limited to the embodiments described above, and various modifications are possible. The embodiments described above are exemplified to explain the present invention in an easy-to-understand manner, and are not necessarily limited to those having all the configurations described. Furthermore, it is possible to replace a part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. Further, it is possible to delete a part of the configuration of each embodiment, or add or replace other configurations. Furthermore, the control lines and information lines shown in the figures are those considered necessary for explanation, and do not necessarily show all the control lines and information lines necessary on the product. In reality, almost all components may be considered to be interconnected. Possible modifications to the above embodiment include, for example, the following.

(1) In each of the above embodiments, an example has been described in which the inclination angle of the central axis of the welding torch 62 with respect to the horizontal plane is applied as the "torch inclination angle". However, the torch inclination angle is not limited to this, and includes the inclination angle of the welding torch 62 with respect to the surface of the welded material 70 at the welding location, the inclination angle of the welding torch 62 with respect to the normal direction of the welded material 70 at the welding location, The inclination angle of the welding torch 62 with respect to the welding direction may be used as the torch inclination angle.

(2) Since the hardware of the control device 100 in each of the above embodiments can be realized by a general computer, information such as the flowcharts shown in FIGS. 3 and 6, and programs, tables, files, etc. that execute the various processes described above. can be stored in a memory, a recording device such as a hard disk, an SSD (Solid State Drive), or a recording medium such as an IC (Integrated Circuit) card, an SD card, or a DVD (Digital Versatile Disc). Therefore, these programs and the like may be stored in these storage media or distributed via transmission lines.

(3) The processes shown in FIGS. 3 and 6 and other processes described above are described as software processes using programs in the above embodiment, but some or all of them can be implemented using ASIC (Application Specific Integrated Circuit). It may also be replaced by hardware processing using an IC for a specific application) or an FPGA (Field Programmable Gate Array). Furthermore, the work evaluation database 121 and reference database 131 shown in FIG. 1 may be provided in a cloud on a network (not shown).

50 Welding worker 62 Welding torch 64 Reflective material application marker (marker)
100 Control device (welding work evaluation device)
110 Coordinate measurement unit (coordinate measurement process, coordinate measurement means)
120 Algorithm selection unit (algorithm selection process, algorithm selection means)
122 Comparison selection unit 123 Posture estimation unit 130 Welding work evaluation unit (welding work evaluation means, welding work evaluation process)
132 Comparison section 134 Database complementation section (complementation section)
150-156 Motion capture device

Claims (7)

a coordinate measuring unit that obtains a torch position that is the position of a welding torch during welding work;
an algorithm selection unit that selects one of the work evaluation algorithms that corresponds to the movement of the torch position from among the plurality of work evaluation algorithms;
a welding work evaluation unit that evaluates the welding work based on the selected work evaluation algorithm ,
The welding work evaluation department
Selecting at least two of the reference data from among a plurality of reference data each corresponding to a part of the plurality of work evaluation algorithms, and interpolating the selected at least two of the reference data to evaluate the selected work. Equipped with a complementation unit that generates new reference data corresponding to the evaluation algorithm
A welding work evaluation device characterized by: The welding torch or the welding operator who operates the welding torch is equipped with a marker,
The coordinate measurement unit has a function of calculating the torch position and a torch inclination angle, which is the inclination of the welding torch, based on information from a motion capture device that measures the position of the marker. The welding work evaluation device according to claim 1. The algorithm selection unit is
a posture estimation unit that estimates a posture of a welding worker operating the welding torch based on the torch position and a torch inclination angle that is a tilt of the welding torch;
The welding work evaluation device according to claim 1, further comprising a comparison selection unit that selects the work evaluation algorithm corresponding to the estimated welding posture that is the estimation result of the posture estimation unit. The welding work evaluation department
One of the reference data corresponding to the selected work evaluation algorithm is selected from among the plurality of reference data corresponding to the plurality of work evaluation algorithms, and the movement of the torch position is determined based on the selected reference data and the movement of the torch position. The welding work evaluation device according to claim 1, further comprising a comparison section for comparing. The welding work evaluation department
The welding work evaluation device according to claim 1 , further comprising a comparison unit that compares the new reference data and the movement of the torch position. a coordinate measurement process for obtaining a torch position, which is the position of a welding torch during welding work;
an algorithm selection step of selecting one of the work evaluation algorithms that corresponds to the movement of the torch position from among the plurality of work evaluation algorithms;
a welding work evaluation step of evaluating the welding work based on the selected work evaluation algorithm;
make the computer run
The welding work evaluation process includes:
Selecting at least two of the reference data from among a plurality of reference data each corresponding to a part of the plurality of work evaluation algorithms, and interpolating the selected at least two of the reference data to evaluate the selected work. Equipped with a complementary process that generates new reference data corresponding to the evaluation algorithm
A welding work evaluation method characterized by: computer,
Coordinate measuring means for obtaining the torch position, which is the position of the welding torch during welding work;
Algorithm selection means for selecting one of the work evaluation algorithms corresponding to the movement of the torch position from among the plurality of work evaluation algorithms;
welding work evaluation means for evaluating the welding work based on the selected work evaluation algorithm;
It is a program to function as
The welding work evaluation means includes:
Selecting at least two of the reference data from among a plurality of reference data each corresponding to a part of the plurality of work evaluation algorithms, and interpolating the selected at least two of the reference data to evaluate the selected work. Equipped with a complementary function to generate new reference data corresponding to the evaluation algorithm
A program characterized by:

Images